Yeah, that's right. What is necessary for the server market is to have the fastest and best specs in every department. Bus speed, processor speed, cache size, the lot. Servers are just like workstations and gaming PCs, after all.
Heat? Component cost? Stability? Never mind. What's important for OLTP and web servers is to have each processor running on the very fastest memory bus in the world.
Coming soon to slashdot: Doom 3 FPS benchmarks for 2U rackmount servers...
First, the lead-in on Slashdot is silly. Intel has been planning this for a while. Yes, it is good customer service. They're not "cutting their losses" and this move makes sense even if AMD were to fold up tomorrow.
It's just simply too expensive to develop two different motherboard sets when you could leverage the increasingly similar characteristics of high-end Xeon motherboards and Itanium motherboards.
Also silly is the end of the article suggesting that Itanium will take over the world any time in the reasonably distant future. This is a strawman will no doubt ignite a frenzy of Itanium hating from various people (yes, we've all heard it before, Itanium is dead, Linus hates it, etc.). Besides, it doesn't reflect Intel's current clearly stated strategy, which indicates that we'll have both architectures for a very long time.
Obviously some of this article is pure FUD. However, I think some of the points are extremely valid. I don't think that the open source world has the same sort of reliability standards as the reliable embedded systems world. All that cheerleading about millions of eyeballs going over open systems code doesn't fix this.
The sort of verification work required to certify code for high-reliabilty systems is very, very boring. I don't know that this sort of work is scratching anyone's personal itch in the way that, say, fixing a broken PC device driver fixes someone's itch.
Another problem is that being the first mover is a huge disadvantage. If you spend tens of millions verifying your source code up to some known standard, you're in the interesting spot of having to release it later. Granted there might be some user-level stuff that you keep out of GPL. Aside from that the rest of the world gets to see a kernel that is known to have passed some expensive, complicated verification standard, and can start there.
I'm not saying that it's impossible for embedded Linux to meet these sort of standards, or that some company somewhere won't find a good reason to do this work even if others benefit from it. But please, spare us the "millions of eyeballs checking the wonderful open source, line by line" stuff. This is far from the level of accountability and detail required from high-reliability systems - even if you have some sort of faith that code had been really well checked over by the open source world, how would you know for sure?
Ummn, no. If you want a nice probabilistic guarantee of security, then you would prefer the wide, well-lit, well-trafficked street.
However, you will note that the White House is not secured by making sure that all of the approaches to it have vibrant, 24/7 goings on so that some concerned citizen will call 911 if they see an assassin heading for the Oval Office.
This assumes all ideas are created equal. It's like saying "I wasn't really all that interested in politics until I read that SF book where the hero defends himself against a teen street gang and has to flee to a Mars colony, where men live Strong and True. Damn the coddling welfare system and perpetrator-oriented justice system. Now, based on that evidence, I'm a Libertarian".
In other words, you only decided to be interested in ecology once someone told you what you wanted to hear in a work of fiction. This isn't an interest in ecology, it's an interest in having your prejudices confirmed.
Also pressing: debate the morality of travelling back in time to impregnate your own grandma, the environmental consequences of switching on the immortality gene, and the ethical issues of putting "off" switches on machines running God-like Artifical Intelligences.
Talk quickly, people. If these developments happen before they are rigorously discussed on Slashdot the future of the planet is in peril.
Absolutely astonishing. I love Slashdot to keep me reminded of how science fiction can "interest" people in the real issues...
That is, someone doesn't like the tone of any environmental writer he's heard so far (using a ludicrous generalization to dismiss all ecological thought), and he supposes that he's now 'interested' in ecology because of some fictional events in a sci-fi book on Mars.
Not sure you're phrasing this as a disagreement - I didn't say we'd solve these problems by being nice and moving to some sort of World Hippy Utopia.
Re:No point in trying to get off Earth (now)
on
The Wrong Stuff
·
· Score: 1
Ah, that was good for a laugh.
"If there were only 200 million humans on Earth, there wouldn't be wars."
Remarkable! Presumably all those accounts of war pre-600 A.D. are ingenious fakes... because that's about the point that world population crossed the 200 million line, depending on who you believe (other numbers say around 1 A.D.)
Famine and poverty have almost nothing to do with population pressure. Some of the most successful, fast-growing countries in the world (the more successful S.E. Asian countries, for example) have immense population pressure and are effectively ignoring all these neo-Malthusian arguments. Meanwhile, many of the poorest African countries have grossly underutilized resources and low population pressure. Sorry, but everything you think you know on this issue is wrong. Superficially attractive, but wrong. I'm always amazed at how little people who push these Malthusian arguments actually know about poor countries; generally these countries are poor because of systemic problems that lead to extremely low productivity per capita, not because they're so heavily populated that no-one has enough to eat.
Your argument would hold if food (and other resources) was a constant that was just distributed over the land, and could be gathered with roughly the same effect regardless of technological and economic sophistication. Fortunately, that's not true.
No point in trying to get off Earth (now)
on
The Wrong Stuff
·
· Score: 5, Insightful
I have very mixed feelings about this. I think Weinberg is basically correct on the issues. But I do feel that at some point, the continuity of the human race may depend on not having all it's eggs in one basket.
However, there's a huge "but" that follows that last paragraph. Right now, putting huge amounts of resources into some sort of manned spaceflight is ridiculous from a scientific perspective and offers no real lessons on ultimately how we're going to do some sort of sustainable long-distance space flight. What, we can't manage to get Biodome working, and we're supposedly going to have Mars colonies?
Putting on my futurist cap for a moment (much like a dunce cap with the advantage that no-one notices it) I'd have to say that there are three major alternatives for how things will develop overall.
1. We all are wiped out in the shortish term by { global warming, killer viruses, giant asteroid, the covering of all Earth's arable land with AOL disks, etc}. In this situation, manned space flight might add a couple artifacts for the alien archeologists to ponder, but isn't going to matter.
2. We see a continued explosion of new technologies in the areas biotech, advanced physics, computing, etc. In this case, why try manned space flight now? What are we going to learn from pushing 1990s technology to eke out a single, unsustainable dash to Mars and back, when there are so many other interesting problems that can drive science. The money would be better spent elsewhere.
3. We neither wipe ourselves out nor see a explosion of new technology; instead, the rate of change goes down and we converge on a pleasant, fairly quiet future. Moore's law comes to an end, biotech doesn't turn out to produce amazing new developments affordably, modern physics offers no real practical advances in day-to-day life. Perhaps gradual technical progress and the dissemination of technologies to the third world makes Earth a nice, comfortable planet. But ultimately, things in 2200 look pretty recognizable to someone from 1980.
In this case, we'll never manage to achieve the sorts of technologies required to leave the solar system or set up a long-term presence anywhere else. I think this future is to some extent the most interesting because no-one takes it seriously - the assumption of a lot of people is that just because we've seen an incredible explosion of new technologies since the industrial revolution, this explosive growth of knowledge and expertise will continue forever.
That may be true - but I think many people (particularly Slashdot nerds) would benefit from thinking about alternative courses. Maybe we'll have to solve all those human-scale problems (war, enivronmental destruction, poverty, disease, human suffering, etc.) without the benefit of self-replicating autonomous nanotech, real AI, faster-than-light travel, Dyson spheres and all those great science-fictional constructs.
To some extent, I think science fiction provides the secular humanist's equivalent of the Rapture. When asked about an enviromental issue, Reagan's secretary of the interior (James Watts, I believe) reponded that he didn't think the environment is such a big deal because this might the last generation before the Rapture. Listening to futurists, I often get the same kind of feeling that they think that today's issues are about as important as an industrial dispute among buggy whip makers in 1910.
I'm just going to ignore the usual rubbish you trot about how anyone who prefers the C/C++ way of doing things has obviously never tasted the joys of the lambda calculus, LISP, Scheme, ML 2000, object-oriented Welsh Left-Handed Haskell and so on. Uh, no.
You're just describing your favorite languages with some handwaving so that it sounds like you've unearthed overarching principles about how programming languages could be so much better. But all you're doing is reeling out a long list of arbitrary preferences. Enjoy them; if nothing else, they obviously give you much to write about on Slashdot.
How does this get modded "insightful"? I'm sort of tired of people parading around their ignorance about programming disguised as 'good design sense'.
"Type-safety is like training wheels."
Yeah, right. One day you graduate to the level of Supreme Programming God, with the power to effortlessly remember the exact types of all the "void *" pointers they left lying around. And of course, you don't need type safety when interfacing to other people's code, because you're just so damn intelligent, you instantly understand everything about their code and don't need those 'training wheels'. Because lord knows we all understand exactly what the other programmers mean and don't need no steenking type system to communicate what should and shouldn't be legal.
"Unfortunately, we've got languages with templates, boxing/unboxing, exceptions, RTTI, etc as our main tools, that often provokes us to design systems that are more complicated than they should be."
I love Slashdot for these kind of broad-brush statments. Which language uses exceptions or RTTI as its 'main tool', exactly? And who the hell thinks exceptions and templates are more complicated than implementing the same features _without_ exceptions or templates.
My guess: someone who has never written a large program* or a program that actually requires generic programming or sophisticated exception handling (yes, sometimes it is OK to just print an error message and quit). The stuff about type safety also suggests near-total lack of experience with programming large systems with portions of the system written by other people.
Also, absent turning newlines into spaces, how do you write a properly formatted 1-line "Hello World" program in C? If you're going to complain about Java's main declaration and System.out, you should at least take into account C's main declaration, and the equally inscrutable (to a first-year student) '#include' line.
* Pre-emptively: to the people that always post about how you don't actually need to ever build large systems, because you can always build elegant collections of small, gem-like, perfect tools, please close your Web browsers and finish your sophomore CS homework.
C++ seems to get mistreated in these debates; the picture of the language that people seem to have is an complete strawman. If you believed the Java/C# crowd, someone puts a gun to your head in C++ and insists that you
a) build massive type hierarchies with all sorts of cryptic overloads and exciting multiple inheritance paradoxes,
b) regularly cast things in and out of void * and use at least three unions per C file, as well as including a large number of K&R prototypes just for the hell of it,
c) use the worst possible C++ compilers available, and completely avoid any of the smart pointer implementations, C++ garbage collectors or tools like Purify that might help you with memory lifetime issues,
d) overload everything in sight, all the time,
e) write the occasional template metaprogram that generates 20K long error messages when you accidentally mess up a type in the splay tree of types implemented entirely in the language type system, and
f) utterly ignore any useful part of C++ that isn't directly expressible in Java or C#, or find compelling reasons to ignore the existence of such features - particularly if the feature was ever poorly implemented in any C++ compiler, anywhere.
C++ certainly gives you more opportunities to write bad code. In fact, it allows you to create huge frameworks that pretty much guarantee that not only is the code you write bad, but anyone who tries to interface with your code will also have to write bad code. However, it also allows the creation of good code and excellent frameworks that can get incredible things done. Well-written STL code, for example, is astoundingly terse, clear, efficient, and wonder of wonders, type-safe!
Unfortunately, there seems to be an increasing trend out there of half-educated programmers with the viewpoint: "if I don't understand it, to hell with it"!
Yeah, sure. What a schmuck this Zuckerman fellow must be.
What kind of low-grade moron doesn't know that the SCO lawsuit and an overly liberal regime of granting software patents is the direct pathway to a horrifying, Blade-Runner-style future where gangs of midgets tear the fittings off your police aircar given half a chance?
I think it's very, very important for any show like this to offer detailed depictions of OSS-type issues. These issues should arise every other show at the very least, and possibly feature verbatim quotes of essays by Eric Raymond and Richard Stallman. A major character might take time out from the courtroom scenes, sex and scandal and face the camera and talk for about 10 minutes about the difference between 'free as in speech' and 'free as in beer'.
"I would state that this licensing project represented only a small fraction of my time over the last year and has completely gone away in recent months. "
and
"I am still hoping people dig up some of the more positive projects I have been involved with."
Before getting involved in this whole SCO mess, Mike should have remembered this vital lesson .
"In the meanwhile, we can all spread the word, discuss, debate and brainstorm every nook and cranny of the program here on Slashdot, and give Edwards a shoulder by giving the program every bit of mass-exposure we can."
Yes, I imagine that that will make all the difference. In future years, the touchstone of scientific and engineering excellence will be "Was it discussed, debated and brainstormed on Slashdot or not?"
Whew, now we know. It looks like in-game violence is in fact not the prime and only cause of all variations in levels of violence across cultures after all. So I guess that wraps the debate up.
Although personally, I think a heavy diet of computer gaming makes people pasty, fat, slow, house-bound and unenterprising and thus unlikely to be able to commit spectacular atrocities. I'd love to see the the statistical comparison for crime rates between gamers vs. pro and college athletes, for example...
From what I recall, there are small amounts of dietary creatine in red meat. This is insignificant compared to the amount of creatine taken when supplementing with creatine (although a study might see more dramatic effects starting from a zero dietary intake).
It's also worth mentioning that you can synthesize creatine in your own body, so it's not like anyone is running around with zero creatine.
You know, there are plenty of lame internet posts with this particular problem, but this may be the first book I've seen devoted to the idea that owning a particular brand of computer makes you a particular type of person. The people who think that owning a Mac makes them 'creative' and 'different' need to get over themselves a little (same with the people that think installing Linux automatically makes them an ubergeek).
Enjoy the OS of your choice, but please don't imagine that choosing it was a major life achievement. Edit together a really cool movie on your G4 (or write a kernel patch if you're an ubergeek wannabe) and you're living up to the hype... but until you've actually done something significant, the choice of OS is about as important as the choice of cola.
Note the phrase at the end of the abstract - "Causality Unproven". While I'm sure that they corrected for what factors they could, I think it's at least plausible that various illnesses that might cause you to require more sleep (for example, by reducing the quality of the sleep that you get) might contribute to higher mortality rates amongst those who get more sleep.
So the study hardly offers a prescription for longer life; on the other hand, it's a pretty impressive rebuttal to any idea that people who sleep less than 8 hours are going to have problems as a result.
cheinonen is quite correct. The only way digital is going to look better is when it's showcasing stuff that was originally done digitally; the fact that there's not much in the way of conversion means that you can do a very good job of rendering scenes digitally targetting a digital projector.
However, for those of us who watch movies that contain predominately filmed content, these digital projectors are third-rate. I'm glad to see the whole thing flop, and I'm sure Ebert is too.
Re:Itanium at 1.6 GHz in 2003 ?
on
Intel's Big Chip
·
· Score: 2
As far as I know, the controversy about SPECfp 2000 relates to Sun's compiler, not IBM's. The usual "Your compiler does too well on SPEC" whining, nothing serious.
I think SPEC needs to release a new benchmark set every 6 months, with about 50 really large programs in it each time. Ok, that's a little silly, but otherwise "gaming" the SPECs is essentially inevitable.
Not only is this a strangely old topic on the face of it, it's even older than they claim. Michael Wise at Sydney University (now a research fellow at Cambridge) had a system called "yap" (Yet Another Plague) which did fairly advanced plagiarism detection on electronically submitted assignments that was already old hat when I was a 1st year undergraduate, back in 1990.
The first paper on yap (from 1992) cites related work from 1981 (in SIGCSE) and a series of articles on the system "plague" as far back as 1986.
All of this material (as well as the successor system, YAP3) is easily available on the Web. It would be nice to see people do 10 minutes of research before spewing out Yahoo news stories and Slashdot posts (note: "research" is a slightly different skill to "cutting and pasting press releases").
Slashdot Mirror
Yeah, that's right. What is necessary for the server market is to have the fastest and best specs in every department. Bus speed, processor speed, cache size, the lot. Servers are just like workstations and gaming PCs, after all.
Heat? Component cost? Stability? Never mind. What's important for OLTP and web servers is to have each processor running on the very fastest memory bus in the world.
Coming soon to slashdot: Doom 3 FPS benchmarks for 2U rackmount servers...
First, the lead-in on Slashdot is silly. Intel has been planning this for a while. Yes, it is good customer service. They're not "cutting their losses" and this move makes sense even if AMD were to fold up tomorrow.
It's just simply too expensive to develop two different motherboard sets when you could leverage the increasingly similar characteristics of high-end Xeon motherboards and Itanium motherboards.
Also silly is the end of the article suggesting that Itanium will take over the world any time in the reasonably distant future. This is a strawman will no doubt ignite a frenzy of Itanium hating from various people (yes, we've all heard it before, Itanium is dead, Linus hates it, etc.). Besides, it doesn't reflect Intel's current clearly stated strategy, which indicates that we'll have both architectures for a very long time.
Important difference between programming and NS's writing: in programming, we occasionally edit or even delete functions...
Obviously some of this article is pure FUD. However, I think some of the points are extremely valid. I don't think that the open source world has the same sort of reliability standards as the reliable embedded systems world. All that cheerleading about millions of eyeballs going over open systems code doesn't fix this.
The sort of verification work required to certify code for high-reliabilty systems is very, very boring. I don't know that this sort of work is scratching anyone's personal itch in the way that, say, fixing a broken PC device driver fixes someone's itch.
Another problem is that being the first mover is a huge disadvantage. If you spend tens of millions verifying your source code up to some known standard, you're in the interesting spot of having to release it later. Granted there might be some user-level stuff that you keep out of GPL. Aside from that the rest of the world gets to see a kernel that is known to have passed some expensive, complicated verification standard, and can start there.
I'm not saying that it's impossible for embedded Linux to meet these sort of standards, or that some company somewhere won't find a good reason to do this work even if others benefit from it. But please, spare us the "millions of eyeballs checking the wonderful open source, line by line" stuff. This is far from the level of accountability and detail required from high-reliability systems - even if you have some sort of faith that code had been really well checked over by the open source world, how would you know for sure?
Ummn, no. If you want a nice probabilistic guarantee of security, then you would prefer the wide, well-lit, well-trafficked street.
However, you will note that the White House is not secured by making sure that all of the approaches to it have vibrant, 24/7 goings on so that some concerned citizen will call 911 if they see an assassin heading for the Oval Office.
This assumes all ideas are created equal. It's like saying "I wasn't really all that interested in politics until I read that SF book where the hero defends himself against a teen street gang and has to flee to a Mars colony, where men live Strong and True. Damn the coddling welfare system and perpetrator-oriented justice system. Now, based on that evidence, I'm a Libertarian".
In other words, you only decided to be interested in ecology once someone told you what you wanted to hear in a work of fiction. This isn't an interest in ecology, it's an interest in having your prejudices confirmed.
Also pressing: debate the morality of travelling back in time to impregnate your own grandma, the environmental consequences of switching on the immortality gene, and the ethical issues of putting "off" switches on machines running God-like Artifical Intelligences.
Talk quickly, people. If these developments happen before they are rigorously discussed on Slashdot the future of the planet is in peril.
Absolutely astonishing. I love Slashdot to keep me reminded of how science fiction can "interest" people in the real issues...
That is, someone doesn't like the tone of any environmental writer he's heard so far (using a ludicrous generalization to dismiss all ecological thought), and he supposes that he's now 'interested' in ecology because of some fictional events in a sci-fi book on Mars.
Not sure you're phrasing this as a disagreement - I didn't say we'd solve these problems by being nice and moving to some sort of World Hippy Utopia.
Ah, that was good for a laugh.
"If there were only 200 million humans on Earth, there wouldn't be wars."
Remarkable! Presumably all those accounts of war pre-600 A.D. are ingenious fakes... because that's about the point that world population crossed the 200 million line, depending on who you believe (other numbers say around 1 A.D.)
Famine and poverty have almost nothing to do with population pressure. Some of the most successful, fast-growing countries in the world (the more successful S.E. Asian countries, for example) have immense population pressure and are effectively ignoring all these neo-Malthusian arguments. Meanwhile, many of the poorest African countries have grossly underutilized resources and low population pressure. Sorry, but everything you think you know on this issue is wrong. Superficially attractive, but wrong. I'm always amazed at how little people who push these Malthusian arguments actually know about poor countries; generally these countries are poor because of systemic problems that lead to extremely low productivity per capita, not because they're so heavily populated that no-one has enough to eat.
Your argument would hold if food (and other resources) was a constant that was just distributed over the land, and could be gathered with roughly the same effect regardless of technological and economic sophistication. Fortunately, that's not true.
I have very mixed feelings about this. I think Weinberg is basically correct on the issues. But I do feel that at some point, the continuity of the human race may depend on not having all it's eggs in one basket.
However, there's a huge "but" that follows that last paragraph. Right now, putting huge amounts of resources into some sort of manned spaceflight is ridiculous from a scientific perspective and offers no real lessons on ultimately how we're going to do some sort of sustainable long-distance space flight. What, we can't manage to get Biodome working, and we're supposedly going to have Mars colonies?
Putting on my futurist cap for a moment (much like a dunce cap with the advantage that no-one notices it) I'd have to say that there are three major alternatives for how things will develop overall.
1. We all are wiped out in the shortish term by { global warming, killer viruses, giant asteroid, the covering of all Earth's arable land with AOL disks, etc}. In this situation, manned space flight might add a couple artifacts for the alien archeologists to ponder, but isn't going to matter.
2. We see a continued explosion of new technologies in the areas biotech, advanced physics, computing, etc. In this case, why try manned space flight now? What are we going to learn from pushing 1990s technology to eke out a single, unsustainable dash to Mars and back, when there are so many other interesting problems that can drive science. The money would be better spent elsewhere.
3. We neither wipe ourselves out nor see a explosion of new technology; instead, the rate of change goes down and we converge on a pleasant, fairly quiet future. Moore's law comes to an end, biotech doesn't turn out to produce amazing new developments affordably, modern physics offers no real practical advances in day-to-day life. Perhaps gradual technical progress and the dissemination of technologies to the third world makes Earth a nice, comfortable planet. But ultimately, things in 2200 look pretty recognizable to someone from 1980.
In this case, we'll never manage to achieve the sorts of technologies required to leave the solar system or set up a long-term presence anywhere else. I think this future is to some extent the most interesting because no-one takes it seriously - the assumption of a lot of people is that just because we've seen an incredible explosion of new technologies since the industrial revolution, this explosive growth of knowledge and expertise will continue forever.
That may be true - but I think many people (particularly Slashdot nerds) would benefit from thinking about alternative courses. Maybe we'll have to solve all those human-scale problems (war, enivronmental destruction, poverty, disease, human suffering, etc.) without the benefit of self-replicating autonomous nanotech, real AI, faster-than-light travel, Dyson spheres and all those great science-fictional constructs.
To some extent, I think science fiction provides the secular humanist's equivalent of the Rapture. When asked about an enviromental issue, Reagan's secretary of the interior (James Watts, I believe) reponded that he didn't think the environment is such a big deal because this might the last generation before the Rapture. Listening to futurists, I often get the same kind of feeling that they think that today's issues are about as important as an industrial dispute among buggy whip makers in 1910.
I'm just going to ignore the usual rubbish you trot about how anyone who prefers the C/C++ way of doing things has obviously never tasted the joys of the lambda calculus, LISP, Scheme, ML 2000, object-oriented Welsh Left-Handed Haskell and so on. Uh, no.
You're just describing your favorite languages with some handwaving so that it sounds like you've unearthed overarching principles about how programming languages could be so much better. But all you're doing is reeling out a long list of arbitrary preferences. Enjoy them; if nothing else, they obviously give you much to write about on Slashdot.
How does this get modded "insightful"? I'm sort of tired of people parading around their ignorance about programming disguised as 'good design sense'.
"Type-safety is like training wheels."
Yeah, right. One day you graduate to the level of Supreme Programming God, with the power to effortlessly remember the exact types of all the "void *" pointers they left lying around. And of course, you don't need type safety when interfacing to other people's code, because you're just so damn intelligent, you instantly understand everything about their code and don't need those 'training wheels'. Because lord knows we all understand exactly what the other programmers mean and don't need no steenking type system to communicate what should and shouldn't be legal.
"Unfortunately, we've got languages with templates, boxing/unboxing, exceptions, RTTI, etc as our main tools, that often provokes us to design systems that are more complicated than they should be."
I love Slashdot for these kind of broad-brush statments. Which language uses exceptions or RTTI as its 'main tool', exactly? And who the hell thinks exceptions and templates are more complicated than implementing the same features _without_ exceptions or templates.
My guess: someone who has never written a large program* or a program that actually requires generic programming or sophisticated exception handling (yes, sometimes it is OK to just print an error message and quit). The stuff about type safety also suggests near-total lack of experience with programming large systems with portions of the system written by other people.
Also, absent turning newlines into spaces, how do you write a properly formatted 1-line "Hello World" program in C? If you're going to complain about Java's main declaration and System.out, you should at least take into account C's main declaration, and the equally inscrutable (to a first-year student) '#include' line.
* Pre-emptively: to the people that always post about how you don't actually need to ever build large systems, because you can always build elegant collections of small, gem-like, perfect tools, please close your Web browsers and finish your sophomore CS homework.
C++ seems to get mistreated in these debates; the picture of the language that people seem to have is an complete strawman. If you believed the Java/C# crowd, someone puts a gun to your head in C++ and insists that you
a) build massive type hierarchies with all sorts of cryptic overloads and exciting multiple inheritance paradoxes,
b) regularly cast things in and out of void * and use at least three unions per C file, as well as including a large number of K&R prototypes just for the hell of it,
c) use the worst possible C++ compilers available, and completely avoid any of the smart pointer implementations, C++ garbage collectors or tools like Purify that might help you with memory lifetime issues,
d) overload everything in sight, all the time,
e) write the occasional template metaprogram that generates 20K long error messages when you accidentally mess up a type in the splay tree of types implemented entirely in the language type system, and
f) utterly ignore any useful part of C++ that isn't directly expressible in Java or C#, or find compelling reasons to ignore the existence of such features - particularly if the feature was ever poorly implemented in any C++ compiler, anywhere.
C++ certainly gives you more opportunities to write bad code. In fact, it allows you to create huge frameworks that pretty much guarantee that not only is the code you write bad, but anyone who tries to interface with your code will also have to write bad code. However, it also allows the creation of good code and excellent frameworks that can get incredible things done. Well-written STL code, for example, is astoundingly terse, clear, efficient, and wonder of wonders, type-safe!
Unfortunately, there seems to be an increasing trend out there of half-educated programmers with the viewpoint: "if I don't understand it, to hell with it"!
Yeah, sure. What a schmuck this Zuckerman fellow must be.
What kind of low-grade moron doesn't know that the SCO lawsuit and an overly liberal regime of granting software patents is the direct pathway to a horrifying, Blade-Runner-style future where gangs of midgets tear the fittings off your police aircar given half a chance?
I think it's very, very important for any show like this to offer detailed depictions of OSS-type issues. These issues should arise every other show at the very least, and possibly feature verbatim quotes of essays by Eric Raymond and Richard Stallman. A major character might take time out from the courtroom scenes, sex and scandal and face the camera and talk for about 10 minutes about the difference between 'free as in speech' and 'free as in beer'.
"I would state that this licensing project represented only a small fraction of my time over the last year and has completely gone away in recent months. "
and
"I am still hoping people dig up some of the more positive projects I have been involved with."
Before getting involved in this whole SCO mess, Mike should have remembered this vital lesson .
"In the meanwhile, we can all spread the word, discuss, debate and brainstorm every nook and cranny of the program here on Slashdot, and give Edwards a shoulder by giving the program every bit of mass-exposure we can."
Yes, I imagine that that will make all the difference. In future years, the touchstone of scientific and engineering excellence will be "Was it discussed, debated and brainstormed on Slashdot or not?"
Hey, that's cool. Where else could you get G5s with ECC memory so cheaply?
They do have ECC memory, right? Having been part of a supercomputer....
Whew, now we know. It looks like in-game violence is in fact not the prime and only cause of all variations in levels of violence across cultures after all. So I guess that wraps the debate up.
Although personally, I think a heavy diet of computer gaming makes people pasty, fat, slow, house-bound and unenterprising and thus unlikely to be able to commit spectacular atrocities. I'd love to see the the statistical comparison for crime rates between gamers vs. pro and college athletes, for example...
From what I recall, there are small amounts of dietary creatine in red meat. This is insignificant compared to the amount of creatine taken when supplementing with creatine (although a study might see more dramatic effects starting from a zero dietary intake).
It's also worth mentioning that you can synthesize creatine in your own body, so it's not like anyone is running around with zero creatine.
You know, there are plenty of lame internet posts with this particular problem, but this may be the first book I've seen devoted to the idea that owning a particular brand of computer makes you a particular type of person. The people who think that owning a Mac makes them 'creative' and 'different' need to get over themselves a little (same with the people that think installing Linux automatically makes them an ubergeek).
Enjoy the OS of your choice, but please don't imagine that choosing it was a major life achievement. Edit together a really cool movie on your G4 (or write a kernel patch if you're an ubergeek wannabe) and you're living up to the hype... but until you've actually done something significant, the choice of OS is about as important as the choice of cola.
Note the phrase at the end of the abstract - "Causality Unproven". While I'm sure that they corrected for what factors they could, I think it's at least plausible that various illnesses that might cause you to require more sleep (for example, by reducing the quality of the sleep that you get) might contribute to higher mortality rates amongst those who get more sleep.
So the study hardly offers a prescription for longer life; on the other hand, it's a pretty impressive rebuttal to any idea that people who sleep less than 8 hours are going to have problems as a result.
cheinonen is quite correct. The only way digital is going to look better is when it's showcasing stuff that was originally done digitally; the fact that there's not much in the way of conversion means that you can do a very good job of rendering scenes digitally targetting a digital projector.
However, for those of us who watch movies that contain predominately filmed content, these digital projectors are third-rate. I'm glad to see the whole thing flop, and I'm sure Ebert is too.
As far as I know, the controversy about SPECfp 2000 relates to Sun's compiler, not IBM's. The usual "Your compiler does too well on SPEC" whining, nothing serious.
I think SPEC needs to release a new benchmark set every 6 months, with about 50 really large programs in it each time. Ok, that's a little silly, but otherwise "gaming" the SPECs is essentially inevitable.
Not only is this a strangely old topic on the face of it, it's even older than they claim. Michael Wise at Sydney University (now a research fellow at Cambridge) had a system called "yap" (Yet Another Plague) which did fairly advanced plagiarism detection on electronically submitted assignments that was already old hat when I was a 1st year undergraduate, back in 1990.
The first paper on yap (from 1992) cites related work from 1981 (in SIGCSE) and a series of articles on the system "plague" as far back as 1986.
All of this material (as well as the successor system, YAP3) is easily available on the Web. It would be nice to see people do 10 minutes of research before spewing out Yahoo news stories and Slashdot posts (note: "research" is a slightly different skill to "cutting and pasting press releases").